Build reusable Python libraries with pgdbm for standalone or shared-pool deployments.
npx skills add https://github.com/juanre/pgdbm --skill dual-mode-library
Please help me install this Agent Skill. Skill: dual-mode-library Source: https://github.com/juanre/pgdbm/tree/main/skills/dual-mode-library Command: npx skills add https://github.com/juanre/pgdbm --skill dual-mode-library
This pattern enables building reusable Python libraries that can run standalone with their own database pool or be embedded into a larger application using a shared pool, avoiding code duplication across services.
Standalone mode: provide a connection_string and call initialize, then use library methods and finally close. lib = MyLibrary(connection_string="postgresql://localhost/mydb") await lib.initialize()
await lib.close()
High-intent search queries and answers about installing and using this skill.
Build dual-mode libraries by accepting either a connection_string or db_manager parameter. This lets your library run standalone with its own pool or embed in larger applications using shared pools, with per-schema migrations ensuring isolation in both deployment patterns.
Use per-schema migrations scoped to your library's namespace. This isolation pattern allows multiple libraries to safely share a single pgdbm pool without schema conflicts while maintaining clean, reusable code for PyPI distribution or internal tooling.
Yes. The dual-mode pattern eliminates code duplication by designing libraries to accept connection management from either mode—standalone initialization or external pool injection—while always running migrations and handling cleanup only for pools the library creates.
Libraries should only close connections they own. By tracking whether the pool was created internally (from connection_string) or passed externally (db_manager), cleanup remains safe in shared environments without terminating pools other services depend on.
The dual-mode approach removes the barrier: design libraries to work standalone or embedded without code changes. Accept flexible connection parameters, enforce schema isolation via migrations, and implement ownership-aware cleanup—enabling distribution without forcing deployment patterns on consumers.
FindMalekConfigure Prisma for multiple database providers and set up the client.
juanreCreate a shared pgdbm pool with schema-isolated managers and template queries.
juanreConfigure a standalone PostgreSQL module with its own connection pool and migrations.
juanreSelect pgdbm deployment patterns based on project topology and migration constraints.
